Combined Kernel-Based BDT-SMO Classification of Hyperspectral Fused Images

Table 4

Comparison of classification results under different SMO multiclass classification approaches (zone 1).

Classification approaches	Overall accuracy (%)	Average accuracy (%)	Kappa coefficient	Training time (s)	Test time (s)	Standard deviation	Optimized parameters
OAO-SMO	82.43	80.06	0.7998	24.36	21.35	0.95	,
OAA-SMO	81.09	79.89	0.7875	25.37	24.10	1.85	,
DAG-SMO	81.25	81.29	0.7904	30.42	15.70	1.36	,
ECC-SMO	82.05	82.34	0.7954	31.71	15.05	1.65	,
BDT-SMO	83.46	83.66	0.8195	30.57	13.67	0.65	,